© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 3062
    Prev
    Next
    INDEX
    Explanations

    sequence following keyword

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects document header and section‐heading tokens (e.g. titles, affiliation labels, dates, and other metadata appearing in headings).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Länge
    -0.90
     accès
    -0.85
    accès
    -0.84
     jenen
    -0.83
     Außerdem
    -0.83
    fondo
    -0.79
    🪛
    -0.79
     करे
    -0.78
     mantan
    -0.78
     СЕ
    -0.78
    POSITIVE LOGITS
    ffff
    0.73
     measurement
    0.72
     ader
    0.70
     feast
    0.70
     desi
    0.69
    nax
    0.69
     Connect
    0.69
     Кубка
    0.69
    quear
    0.69
     calculation
    0.67
    Activations Density 0.021%

    No Known Activations